TABLE 1—
Population, Singleton Preterm Birth, and Census Tract Characteristics, by 1938 Home Owners’ Loan Corporation (HOLC) Grade: New York City, 2013–2017
| By HOLC Grade |
||||||
| Variable | Total, No. (%) | A (Green)— “Best,” No. (%) | B (Blue)—“Still Desirable,” No. (%) | C (Yellow)—“Definitely Declining,” No. (%) | D (Red)— “Hazardous,” No. (%) | Not Classified (Gray), No. (%) |
| Individual level | ||||||
| Total singleton births (row %) | 528 096 (100) | 5 466 (1.0) | 73 950 (14.0) | 202 219 (38.3) | 153 460 (29.1) | 93 001 (17.6) |
| Preterm births (% of total singleton births) | 36 902 (7.0) | 274 (5.0) | 4 957 (6.7) | 14 098 (7.0) | 11 263 (7.3) | 6 310 (6.8) |
| Race/ethnicity (column %) | ||||||
| Non-Hispanic White | 165 103 (31.3) | 3 892 (71.2) | 27 064 (36.6) | 50 912 (25.2) | 45 582 (29.7) | 37 653 (40.5) |
| Non-Hispanic Black | 104 134 (19.7) | 98 (1.8) | 13 136 (17.8) | 35 133 (17.4) | 39 103 (25.5) | 16 664 (17.9) |
| Non-Hispanic Asian and Pacific Islander | 90 929 (17.2) | 720 (13.2) | 10 052 (13.6) | 46 640 (23.1) | 16 829 (11.0) | 16 688 (17.9) |
| Non-Hispanic other | 1 952 (0.4) | 14 (0.3) | 192 (0.3) | 886 (0.4) | 539 (0.4) | 321 (0.4) |
| Hispanic | 160 885 (30.5) | 670 (12.3) | 22 774 (30.8) | 67 044 (33.2) | 49 782 (32.4) | 20 615 (22.2) |
| Not provided or missing | 5 093 (1.0) | 72 (1.3) | 732 (1.0) | 1 604 (0.8) | 1 625 (1.1) | 1 060 (1.1) |
| Mother’s education (column %) | ||||||
| < high school | 103 443 (19.6) | 106 (1.9) | 11 614 (15.7) | 45 366 (22.4) | 34 853 (22.7) | 11 504 (12.4) |
| High school or GED | 121 354 (23.0) | 200 (3.7) | 14 091 (19.0) | 53 931 (26.7) | 37 209 (24.2) | 15 923 (17.1) |
| ≥ some college | 301 774 (57.1) | 5 147 (94.2) | 48 069 (65.0) | 102 436 (50.7) | 80 785 (52.6) | 65 337 (70.2) |
| Not provided (missing) | 1 525 (0.3) | 13 (0.2) | 176 (0.2) | 486 (0.2) | 613 (0.4) | 237 (0.2) |
| Census tract level | ||||||
| Census tracts: no. (row %)a | 2 166 (100) | 32 (1.5) | 304 (14.0) | 824 (38.0) | 541 (25.0) | 465 (21.5) |
| Poverty level, % below US poverty line (column %) | ||||||
| Very low (< 10%) | 664 (30.7) | 30 (93.8) | 101 (33.2) | 223 (27.1) | 86 (15.9) | 224 (48.2) |
| Medium (10% to < 20%) | 706 (32.6) | 1 (3.1) | 114 (37.5) | 312 (37.9) | 150 (27.7) | 129 (27.7) |
| High (20% to < 30%) | 412 (19.0) | 1 (3.1) | 52 (17.1) | 167 (20.3) | 142 (26.2) | 50 (10.8) |
| Very high (≥ 30%) | 384 (17.7) | 0 (0.0) | 37 (12.2) | 122 (14.8) | 163 (30.1) | 62 (13.3) |
| Cumulative poverty level for high poverty, > 20% below US poverty lineb (column %) | ||||||
| 0 time periods | 899 (41.5) | 29 (90.6) | 149 (49.0) | 378 (45.9) | 104 (19.2) | 239 (51.4) |
| 1–2 time periods | 449 (20.7) | 2 (6.2) | 71 (23.4) | 210 (25.5) | 95 (17.6) | 71 (15.3) |
| 3–5 time periods | 761 (35.1) | 1 (3.1) | 84 (27.6) | 234 (28.4) | 339 (62.7) | 103 (22.2) |
| Unranked | 57 (2.6) | 0 (0.0) | 0 (0.0) | 2 (0.2) | 3 (0.6) | 52 (11.2) |
| ICE for racialized economic segregationc (column %) | ||||||
| Tercile 1 (best off) | 706 (33.3) | 32 (100.0) | 133 (43.8) | 193 (23.5) | 146 (27.0) | 202 (48.0) |
| Tercile 2 | 707 (33.4) | 0 (0.0) | 89 (29.3) | 370 (45.0) | 132 (24.4) | 116 (27.6) |
| Tercile (worst off) | 706 (33.3) | 0 (0.0) | 82 (27.0) | 259 (31.5) | 262 (48.5) | 103 (24.5) |
| ICE for housing tenured (column %) | ||||||
| Tercile 1 (best off) | 706 (33.3) | 25 (78.1) | 98 (32.2) | 319 (38.8) | 78 (14.4) | 186 (44.2) |
| Tercile 2 | 707 (33.4) | 5 (15.6) | 119 (39.1) | 284 (34.6) | 192 (35.6) | 107 (25.4) |
| Tercile (worst off) | 706 (33.3) | 2 (6.2) | 87 (28.6) | 219 (26.6) | 270 (50.0) | 128 (30.4) |
Note. GED = general educational development; ICE = index of concentration at the extremes. We used 2010 normalized census tract boundaries. See Table A and Table C (available as supplements to the online version of this article at http://www.ajph.org) for technical details on the HOLC grade and census tract variables.
1 tract was missing from the analysis.
Time periods: 1980, 1990, 2000 decennial census, and 2008–2012 and 2013–2017 American Community Survey.
High-income White households vs low-income Black households. Missing from analysis = 48 census tracts.
Homeowners vs renters. Missing from analysis = 48 census tracts.